## Scripts included in the replication archive: 

1. manuscript.R is the replication code for the manuscript

2. helpers.R contains a series of helper functions needed to run the code in appendix_s2_1.R and appendix_s2_1.R

3. appendix_s2_1.R is the replication code for SI Appendix Section S2.1

4. appendix_s2_2.R is the replication code for SI Appendix Section S2.2

Note: you must run manuscript.R prior to appendix_s2_1.R and appendix_s2_1.R as the latter scripts rely on intermediate datasets created by manuscript.R

## Datasets included in the replication archive: 

1. municipal_survey.rds is the merged survey dataset for baseline (R_t0 = 1) and followup waves (R_t1 = 1) of Yonkers residents

2. national_survey.rds is the survey dataset for the national sample of US adults

3. conjoint_stacked.rds is the combined dataset for the conjoint experiments fielded on Yonkers police officers and community residents 

## Supplementary datasets: 

4. combined_long_survey.rds is the combined survey data on misperceptions from the municipal and national samples, stored in long format. This is used to create Fig. 2 and Tables S4-S6

5. polcomrep.rds is the combined LEMAS/Census dataset used to create Fig. 1

6. national_target_marginals.rds population proportions used to create Table S1

7. municipal_target_marginals.rds population proportions used to create Table S2

8. grf_municipal.rds estimates from causal forest estimator (see appendix_s2_1.R) in the municipal sample

9. grf_municipal_test.rds test calibration of causal forest estimator in municipal sample(output from test_calibration() in appendix_s2_1.R)

10. grf_national.rds estimates from causal forest estimator (see appendix_s2_1.R) in the national sample

11. grf_national_test.rds test calibration of causal forest estimator in national sample (output from test_calibration() in appendix_s2_1.R)

